4c9153650f
Cosmetic and efficiency changes.
2013-09-20 23:17:17 +02:00
15825c9034
Changed logic in numgrad2.
...
Do not compute the left derivative if the evaluation of the objective function at x+h returns a non zero cost_flag (ie a problem with BK, steady state, ...).
2013-09-20 23:17:17 +02:00
d116fea0d9
Removed useless routines for gradient evaluation (called by csminwell).
2013-09-20 23:17:17 +02:00
22968497ad
Fixed bug manifesting if a model is first estimated with cova_compute=0 and if the resulting mode file is loaded after.
...
Added condition on the existence of hh in the mode_file.
2013-09-19 15:25:23 +02:00
b95f000d8f
Added new routine to generate strings randomly.
2013-09-19 14:43:35 +02:00
9f63c4081f
Fixed bug affecting fs2000/fs2000_calib.mod and shock_decomp_calibrated_model/example1_calib_shock_decomp.mod.
2013-09-18 11:38:58 +02:00
7dc6013eaa
Added headers.
2013-09-18 11:34:10 +02:00
a51d8bfe31
Added Baxter and King (1999) band pass filter for dynSeries objects.
2013-09-18 11:34:10 +02:00
07b50fd8b8
Added missing tag for triggering unitary tests.
2013-09-18 11:34:10 +02:00
52e5e44b1c
Do not produce plot in dynSeries::hpcycle unitary test.
2013-09-18 11:34:10 +02:00
ce4fe4113d
Added methods in dynSeries class for filtering (hptrend and hpcycle).
2013-09-18 11:34:10 +02:00
a29db4175d
fixed bug with calibrated models in dyn_estimation_init.m introduced
...
in commit 952139b
2013-09-18 11:23:47 +02:00
ea69f50134
Improve display of posterior parameters and clarify use of HPD
2013-09-16 17:11:25 +02:00
e123670675
Fix bug introduced in 952139b
...
The 'hh' variable is now always defined, so we now test if it is empty.
2013-09-13 18:23:24 +02:00
743252a571
Fixed typo.
...
Solves failing tests for dynSeries' constructor reported here:
http://www.dynare.org/stepan/dynare/tests/report-de590fb2f7dbe41b25097e12b09037a595f1b6d0.log
2013-09-12 10:56:05 +02:00
de590fb2f7
bug fix: include nofunctions in global_initialization.m
2013-09-11 14:09:22 -04:00
0ed35166de
Added unitary tests (instantiation of the dynSeries class form an excel file).
...
(cherry picked from commit 84075cab81
2013-09-11 17:25:07 +02:00
109066de2b
Added the possibility to instantiate the dynSeries class from an excel (xls or xlsx) file.
...
Closes #348
(cherry picked from commit b7b4a9afdc
2013-09-11 17:24:12 +02:00
952139bbc9
Changes related to the treatment of the (optionally) provided mode file.
...
(1) Added more checks on the content of the provided mode file compared the the list of declared parameters (the condition on the number of parameters is not strong enough).
(2) Added a mechanism to adapt the content of the mode file if possible. For instance, if the estimated parameters are a subset of the parameters in the mode file, we only need to discard some of the parameters in the mode file.
(3) Added output argument in dynare_estimation_init, which returns the hessian matrix (hh) with the estimated mode.
2013-09-10 22:25:16 +02:00
afed204526
Cosmetic. Renamed hh (figure handle not to be confused hessian matrix) as fh.
2013-09-10 22:25:16 +02:00
ea5ee4722b
Reintroduces nofunctions option mentioned in the manual that got somehow lost
2013-09-10 16:35:50 +02:00
2e2315611c
Fixed bug (conversion of variable names in tex).
...
(cherry picked from commit 80218a1b65
2013-09-09 12:16:19 +02:00
f337c2b072
Fixed bugs (misuse of inputname function, as in commit #4ce2c3644af333733d802383559de8c0340213bd).
2013-09-07 14:31:50 +02:00
6b237eb6c2
Added method dynTimeIndex::mpower -> lead-lag operators à la G-M ;-).
...
This method can be used to apply the lead and lag methods an arbitrary number of times to a dynSeries object. For instance, if ts is a dynSeries object, and if we define
>> B = dynTimeIndex()-1;
>> F = dynTimeIndex()+1;
B and F can be used as lag and lead operators and the following syntax:
>> us = ts(F^2);
is equivalent to
>> us = ts.lead(2)
or
>> us = ts.lead.lead
2013-09-06 16:49:40 +02:00
f14db386c6
Added missing case for set_names in subsref. This commit and the previous one fix the bug in @dynSeries::set_names method reported at http://www.dynare.org/stepan/dynare/tests/report-f4c01a43bc2882e7fd18f3cd82344fff9644088c.log .
2013-09-06 16:27:25 +02:00
e641cdd455
Fixed unitary tests. If not specified, the name a variable in a dynSeries object is not '--NaN--' but 'Variable_xx' where x is the number of the variable.
2013-09-06 16:27:25 +02:00
4ce2c3644a
Fixed bug (The argument of inputname must be the input argument number not the input argument itself).
2013-09-06 12:01:53 +02:00
23a6ad22dd
Fixed bug causing unitary tests for dynSeries::plus and dynSeries::minus to fail.
...
See http://www.dynare.org/stepan/dynare/tests/report-f4c01a43bc2882e7fd18f3cd82344fff9644088c.log
2013-09-06 10:33:50 +02:00
f4c01a43bc
Introduce new syntax for shifting backward or forward a dynSeries object.
...
Suppose that ts is a dynSeries object and let
t = dynTimeIndex();
Then
us = ts.lag(); <=> us = ts(t-1);
us = ts.lag(2); <=> us = ts(t-2);
us = ts.lead(); <=> us = ts(t+1);
Added corresponding unitary test.
2013-09-05 17:46:04 +02:00
fecb61421f
Added new class (dynTimeIndex).
2013-09-05 17:38:51 +02:00
feff44fe79
Added dynSeries::lead method.
2013-09-05 16:51:10 +02:00
1edb4c6d9a
Added condition on dynSeries::lag method (input must be positive).
2013-09-05 16:50:27 +02:00
2975350319
Fixed typo (cosmetic).
2013-09-05 16:49:44 +02:00
c9c0a25538
Added missing semicolon and deleted trailing whitespaces.
2013-09-05 12:31:37 +02:00
12fc7a4241
Cosmetic changes. Removed useless curly braces and deleted trailing white spaces.
2013-09-05 12:17:55 +02:00
02eb19fb49
Fixed bug. Added missing case for calls to the dynSeries::lag method.
2013-09-05 12:16:07 +02:00
b76b0c8add
Moved the commented first line, used to declare that the routine has unitary tests, after the declaration of the function, otherwise help is not displayed.
2013-09-05 11:34:19 +02:00
61c4244410
Added unitary test.
2013-09-05 10:14:02 +02:00
56be6eb463
Added new interface to dynDate. The constructor accepts a character ('Y', 'A', 'Q', 'M' or 'W') as the second input to specify the frequency instead of an integer (1, 4, 12, or 52). The following expressions are equivalent:
...
a = dynDate([1945 3],4);
a = dynDate([1945 3],'Q');
2013-09-05 10:14:02 +02:00
d44732a8ca
Fixed copy/paste typo in unitary test.
2013-09-05 10:14:02 +02:00
9a5749343c
Added tests in dynDate constructor (when dynDate is called with two input arguments: time and frequency).
2013-09-05 10:14:02 +02:00
a2d5873a3f
Fixed typo (useless space).
2013-09-05 10:14:02 +02:00
c47131f052
Cosmetic change.
2013-09-05 10:14:02 +02:00
ce400706ef
Added test for integer input (dynDate constructor with one numeric input).
2013-09-05 10:14:02 +02:00
0f10880d07
Added 'Y' or 'A' for yearly data in dynDate class.
...
Follwing expressions are equivalent:
a = dynDate('2009');
a = dynDate('2009y');
a = dynDate('2009a');
a = dynaDate(2009);
2013-09-05 10:14:02 +02:00
4dced05d1d
Fixed typo.
2013-09-05 10:14:02 +02:00
d0fed77f97
Added routine for testing if a string contains letter(s).
2013-09-05 10:14:02 +02:00
6183679fd9
Merge pull request #465 from JohannesPfeifer/conditional_variance_decomp
...
Bugfix for conditional variance decomposition
2013-09-04 02:21:40 -07:00
c2e6feeb75
Deletion of unused definition of H_is_diagonal
...
Cosmetic fix
2013-09-04 11:11:29 +02:00
e1a0fd0524
Bugfix for conditional variance decomposition
...
sigma_e_is_diagonal was erroneously set to 0 if the covariance matrix was diagonal, but contained a 0 on the main diagonal.
2013-09-04 11:11:14 +02:00
Stéphane Adjemian (Charybdis)
Michel Juillard
Johannes Pfeifer
Sébastien Villemot
Houtan Bastani
Sébastien Villemot